1: \expandafter\ifx\csname qd\endcsname\relax
2: \documentstyle[12pt,twoside]{article}\input{cbmacro}\fi
3: \expandafter\ifx\csname ok\endcsname\relax
4: \def\re{{\rm Re}} \def\im{{\rm Im}}
5: {\catcode`@=11 \@addtoreset{equation}{section}\@addtoreset{figure}{section}}
6: \def\theequation{\thesection .\arabic{equation}}
7: \def\thefigure{\thesection .\arabic{figure}}
8: \addtolength{\topmargin}{-48pt}
9: \addtolength{\textheight}{90pt}
10: \addtolength{\evensidemargin}{-31pt}
11: \addtolength{\oddsidemargin}{-12pt}
12: \addtolength{\textwidth}{30pt}
13: \addtolength{\footskip}{21pt}
14: \setlength{\parindent}{0pt}\frenchspacing
15: \begin{document}
16: \fi
17: %\addtocontents{toc}{\protect\vspace*{\tocsp}}
18: \section[Schr\"odinger picture field theory in k=0
19: Friedmann-Robertson-Walker\protect\\ spacetimes]
20: {Schr\"odinger picture field theory in k=0\protect\\
21: Friedmann-Robertson-Walker spacetimes}
22: \label{FRWQFT}
23: This section deals with the quantum theory of a free, massive scalar field in
24: a spatially flat (k=0) Friedmann-Robertson-Walker (FRW) spacetime, and it is
25: a brief summary of some results from reference \cite{QFT}.\\
26: Since dimensional regularization will be applied later on, we work in $d{+}1$
27: spacetime dimensions and on flat $d$-dimensional spacelike hyperplanes with
28: coordinates $\1x=(x_1,\ldots,x_d)$. The maximum symmetry of these hyperplanes
29: will greatly simplify the following calculations. The k=0 FRW-metric has the
30: form
31: \beq ds^2 = dt^2 - a^2(t)\,d\vec{x}\cdot d\vec{x} = g_{\mu\nu}dx^{\mu}dx^{\nu},
32: \label{Met}\eeq
33: where $a(t)$ is the FRW scale factor. In terms of the Hubble function
34: $H(t):=\9a(t)/a(t)$, $\9a:=\6_ta$, one obtains for the Ricci tensor
35: \beq R_{00}=-d(\dot{H}+H^2)\;,\qqd R_{ij}=a^2(\dot{H}+dH^2)\de_{ij}\;,\qqd
36: R_{0i}=0 \eeq
37: and for the curvature scalar $R=-d(2\dot{H}+(d+1)H^2)$.\\
38: A scalar field $\phi$ of mass $m$ is supposed to interact only with the
39: classical gravitational field $g_{\mu\nu}$ and may have an arbitrary coupling
40: $\xi$ to the curvature scalar $R$. Its action reads ($\sqrt{g}:=a^d$)
41: \beq S=\int\!d^{d+1}\!x\,\sqrt{g}\,\2\,(g^{\mu\nu}(\6_{\mu}\phi)(\6_{\nu}\phi)
42: -(m^2\!-\!\xi R)\phi^2). \label{Wirk}\eeq
43: The quantum theory is formulated in the Schr\"odinger picture using a wave
44: functional to represent the quantum state. This shows very clearly
45: the real time evolution character of our analysis. Then the quantum operators
46: are acting in the Fock space on wave functionals $\Psi[\phi;t]$.\\
47: The spacelike derivatives contained in the Hamiltonian can be dealt with by
48: performing a Fourier transform ($d\8k:=d^d\!k/(2\pi)^{d\!/\!2}$,
49: $\a^*(\1k)=\a(-\1k)$):
50: \beq \phi(\1x)\,=\,\int d\8k\,e^{i\1k\1x}\,\a(\1k)\;,\qqd
51: {\de\0\de\phi(\1x)}\,=\,\int d\8k\,e^{-i\1k\1x}\,{\de\0\de\a(\1k)}\;,
52: \label{Fourier}\eeq
53: and the Schr\"odinger equation resp. the Hamiltonian takes the following form:
54: \bea i\,\6_t\,\Psi[\a;t]\!&=&\!\hat{H}\,\Psi[\a;t] \label{Schroe}\\
55: \hat{H}\!&=&\!\2\int d^d\!k\,\Bigl(-{1\0\sqrt{g}}
56: {\de^2\0\de\a(\1k)\de\a(-\1k)}+\sqrt{g}(a^{-2}\1k^2\!+\!m^2\!-\!\xi R)
57: \a(\1k)\a(-\1k)\Bigr) \nn\eea
58: For a curved
59: spacetime without an everywhere timelike Killing vector field no unique Fock
60: vacuum does exist. Rather there is a whole class of Fock vacua, which can all
61: be represented by Gaussian wave functionals. We want our $\Psi[\a;t]$ to be
62: a member of this class. Furthermore we require the quantum state
63: not to break spontaneously the symmetries of the k=0 FRW metric (homogeneity
64: and isotropy of the spacelike hyperplanes), which leads to the following
65: wave functional parametrized by one function $A(k,t)$ (the inverse Gaussian
66: width):
67: \beq \Psi[\a;t]\;=\;N(t)\exp\Bigl(-\2\int d^d\!k\,A(k,t)\,\a(\1k)\,\a(-\1k)-i\Om(t)
68: \Bigr), \label{Gaus}\eeq
69: where $N(t)$ is a real normalization factor, $\Om(t)$ a real phase and
70: $k:=|\1k|$.\\
71: Substituting (\ref{Gaus}) in the Schr\"odinger equation (\ref{Schroe}) we get the
72: equation of motion for $A(k,t)$:
73: \beq i\,\9A(k,t)\;=\;{A^2(k,t)\0\sqrt{g(t)}}
74: -\sqrt{g(t)}\,(a^{-2}(t)k^2\!+\!m^2\!-\!\xi R(t)) \label{Bew1}\eeq
75: This is Riccati's equation, and by the transformation
76: $A(k,t)=:\sqrt{g}(\G(k,t)+i{d\02}H(t))$ it takes the standard form
77: \beq i\,\9{\G}(k,t)\;=\;\G^2(k,t)+{d^2\04}H^2(t)+{d\02}\9H(t)
78: -(a^{-2}(t)k^2\!+\!m^2\!-\!\xi R(t)), \label{Bew2}\eeq
79: which can be converted by $\G(k,t)=:-i\6_t\ln u(k,t)$ into the linear equation
80: \beq \ddot{u}\;-\;({d^2\04}H^2+{d\02}\9H-(a^{-2}k^2\!+\!m^2\!-\!\xi R))\,u\;=\;0
81: \qd. \label{Bew3}\eeq
82: From (\ref{Bew1}) the following equations for $A(k,t)$ can be derived, which
83: are useful for the calculation of the energy momentum expectation value:
84: \beq -\6_t\,\lb({1\02\,\re A}\rb)\;=\;{1\0\sqrt{g}}\,{\im A\0\re A}
85: \label{Bew4}\eeq
86: \beq {1\0g}\,{|A|^2\02\,\re A}\;=\;(a^{-2}k^2\!+\!m^2\!-\!\xi R)\,{1\02\,\re A}
87: +{1\02\,\sqrt{g}}\,\6_t\lb(\sqrt{g}\6_t\lb({1\02\,\re A}\rb)\rb)\label{Bew5}\eeq
88: The energy momentum tensor acting as the source in the Einstein equations
89: is defined as variational derivative of the matter action with respect to the
90: metric tensor
91: \[ T_{\mu\nu}(x)\;:=\;{2\0\sqrt{g}}\,\,{\de S\0\de g^{\mu\nu}(x)}\;.\]
92: Due to the spatial symmetries the expectation value of the corresponding
93: operator in the Gaussian state can be written as
94: \beq \<\Psi|({T^{\mu}}_{\nu})|\Psi\>\;=\;\left(\begin{array}[c]{cccc}
95: \r(t)\\&-p(t)\\&&-p(t)\\&&&-p(t)\end{array}\right)\;,\eeq
96: and the explicit calculation leads to the energy density $\r$ and pressure $p$
97: in terms of the width $A(k,t)$:
98: \bea \r\!&=&\!
99: \2\int{\8d\8k\02\,\re A(k,t)}\biggl({|A(k,t)|^2\0g}+a^{-2}k^2+m^2
100: +2\xi\, G_{00}-2\xi\,{dH\0\sqrt{g}}\,2\,\im A(k,t)\biggr) \nn\\
101: p\!&=&\!
102: \2\int{\8d\8k\02\,\re A(k,t)}\biggl({|A(k,t)|^2\0g}
103: +({2\0d}\!-\!1)\,a^{-2}k^2-m^2+2\xi\, a^{-2}G_{11} \nn\\
104: &&\!\qd+\;4\xi\,(a^{-2}k^2\!+\!m^2\!-\!\xi R)-4\xi\,{|A(k,t)|^2\0g}-2\xi\,
105: {H\0\sqrt{g}}\,2\,\im A(k,t)\biggr) \label{EIT4}\eea
106: $G_{\mu\nu}=R_{\mu\nu}-\2\,g_{\mu\nu}\,R$ is the Einstein tensor,
107: and $\8d\8k:=d^d\!k/(2\pi)^d$.\\
108: Equation (\ref{Bew1}) shows, that $A(k,t)$ is of the order of $k$ for large $k$.
109: Hence the energy momentum expectation value (\ref{EIT4}) is quartic divergent
110: and has to be renormalized. These ultraviolet divergencies are due to the
111: behaviour of the wave functional for field configurations of high energy and
112: momenta (large $k$) or resp. for small distances and are connected to the local
113: geometry of the underlying spacetime manifold. For this reason they should be
114: proportional to local geometric tensors which can be absorbed into the
115: gravitational part of the Einstein equations. Thus the divergencies can be
116: removed by a renormalization of the physical parameters in the Einstein
117: equations (cosmological constant, Newton's constant and additional parameters
118: mentioned below). Fortunately the divergencies of the energy momentum
119: expectation value can be calculated as a local functional of a general metric
120: tensor by means of the De~Witt-Schwinger-Christensen expansion.
121: It turns out that in the Einstein equations one has to admit
122: the geometrical tensors $H_{\mu\nu}$, $^{(1)}\!H_{\mu\nu}$ and
123: $^{(2)}\!H_{\mu\nu}$, which are the metric variations
124: $1/\sqrt{g}\;\de/\de g^{\mu\nu}$ of the functionals
125: $\int d^{d\!+\!1}\!x\,\sqrt{g}\,R^{\a\be\r\s}R_{\a\be\r\s}$,
126: $\int d^{d\!+\!1}\!x\,\sqrt{g}\,R^2$ and
127: $\int d^{d\!+\!1}\!x\,\sqrt{g}\,R^{\a\be}R_{\a\be}$.
128: Their renormalized coefficients have to be regarded as additional physical
129: parameters of the theory.
130: We will choose them to be zero, since the effects of these terms have already
131: been analyzed elsewhere \cite{Kuss}.\\
132: The renormalization scheme consists in a subtraction
133: of the first three divergent terms of the De~Witt-Schwinger-Christensen series
134: from the expectation value (\ref{EIT4}):
135: \beq \<T_{\mu\nu}\>_{\rm ren}\;:=\;
136: \<\Psi|T_{\mu\nu}|\Psi\>-\<T_{\mu\nu}\>_{\rm DS\,div} \eeq
137: If $|\Psi\>$ is a state of finite energy density (compared with an adiabatic
138: vacuum as will be explained later on), then the divergencies of
139: $\<\Psi|T_{\mu\nu}|\Psi\>$ and $\<T_{\mu\nu}\>_{\rm DS\,div}$ will cancel
140: and $\<T_{\mu\nu}\>_{\rm ren}$ is finite. It should be noted, that the
141: renormalization scheme decides about the physical meaning of the renormalized
142: parameters.\\
143: With dimensional regularization ($d=3-\e$) one obtains \cite{BirDav}\\
144: \parbox{\textwidth}{
145: \bea \<T_{\mu\nu}\>_{\rm DS\,div}&=&{1\016\pi^2}\,\biggl({1\0\e}-\2\Bigl(\g+
146: \ln{m^2\04\pi}\Bigr)\biggr)\!\cdot\!
147: \biggl({-4\,m^4\0(d\!+\!1)(d\!-\!1)}\,g_{\mu\nu}-{4\,m^2\0d-1}
148: (\xi\!-\!{1\06})G_{\mu\nu}\nn\\
149: & &\qqqd\qqqd+\;{1\090}(H_{\mu\nu}-{}^{(2)}\!H_{\mu\nu})+
150: (\xi\!-\!{1\06})^2\,{}^{(1)}\!H_{\mu\nu}\biggr),\label{TDS2}\eea}
151: where $\g$ is the Euler-Mascheroni constant. For our FRW metric (\ref{Met})
152: the $H$-tensors are explicitly given in appendix \ref{geoT}.
153: \expandafter\ifx\csname ok\endcsname\relax
154: \end{document}\fi
155: